Background

Research has shown that the proper use of a shoulder/lap belt during a crash reduces the risk of death for front seat occupants by 45 percent and the risk of moderate-to-critical injury by 50 percent (Kahane, 2000).  According to the National Highway Traffic Safety Administration’s (NHTSA) Fatality Analysis Reporting System (FARS, 2003), belt use among fatally injured front seat occupants of passenger vehicles declines across the hours of night.

New York passed the Nation’s first seat belt law in 1984.  Belt use among front seat occupants of passenger vehicles increased from about 15 percent before the law to about 50 percent after the law.  However, fatalities in New York decreased by only nine percent; far less than would be expected given the substantial rise in seat belt use.  It was argued that high-crash-risk drivers were less likely to comply with the law than the general driving population.  That is, those most in need of seat belts were least likely to buckle up.  One such high-risk group are those who choose to drink and drive.  Earlier data from checkpoints conducted in Ontario, Canada, showed that only 36 percent of drinking drivers wore their seat belts as compared with 62 percent of non-drinking drivers.  Similar results were reported from British Columbia, Saskatchewan, and Denmark (Lawson et al., 1982; Noordzij et al., 1988).

Preusser et al. (1986) conducted a study to test the theory that very high-risk drivers (e.g., drinking drivers traveling at night) were the least likely to use seat belts.  Preusser et al. tested this theory at and near bars and taverns in New York State six months after the New York seat belt law went into effect.  Observations near taverns showed that 43 percent of drivers passing by during daylight hours were belted, 36 percent of drivers passing by the same locations at night (9 p.m.-2 a.m.) were belted, and just 24 percent of drivers leaving bar parking lots at night were belted. 
Results from the New York bar study clearly indicated that drinking drivers would be a very worthwhile target for a belt use enforcement campaign.  Or, unbelted drivers at night would be a worthwhile target for an impaired-driving enforcement campaign.  These two could be combined into one overall nighttime enforcement effort.  The only known study to combine belt enforcement with alcohol enforcement was the Buckle Up and Drive Sober program in Binghamton, New York, a small upstate community with a self-contained media market (Wells et al., 1986).  The program concept was that belt law violations would be enforced, particularly at night, and each driver stopped would be screened using a passive alcohol sensor.  The effort included a series of combined belt-use and alcohol night checkpoints.  The program, supported with paid media, ran from the fall of 1988 through the fall of 1990.  Belt use was measured through day and night belt use observations.  Drinking and driving was measured using voluntary and anonymous breath testing at checkpoints.  During the course of the program, more than 5,000 drivers were tested for BAC at checkpoints, more than 10,000 drivers were observed for belt use during nighttime hours, and more than 10,000 drivers were observed at the same locations during daylight.  Results indicated that the number of drinking drivers at checkpoints decreased from 23 percent before the program to 14 percent after.  Belt use at night rose from 35 percent to 49 percent. 

Chaudhary et al. (2005b) was the first statewide study to measure seat belt usage at night using a random stratified sampling plan compliant with Federal Register Guidelines for a daytime statewide survey of seat belt usage.  This study included the collection of day and night belt use in Connecticut.  Nighttime procedures were patterned after daytime observation procedures, insofar as possible.  For the first time, a nighttime observational survey made use of night vision technology, enabling observers to make observations in not only well-lit locations, but also non-lit locations.  The results of these surveys indicated higher statewide belt use during daytime hours (83 percent) compared to nighttime hours (77 percent). 

Chaudhary et al. (2005a) was the first study to combine night vision technology with nighttime belt-use enforcement.  The objective of this study was to evaluate the impact of a nighttime seat belt enforcement program.  The program was conducted in Reading, Pennsylvania, a small city (population 81,201) with a self-contained media market.  The program was conducted during September 2004 and only addressed nighttime seat belt use.  Police patrols were equipped with near-military-grade night vision goggles.  Press releases and news stories depicted the goggles as “new” technology available to police.  The program was supported by earned media only.  Bethlehem, Pennsylvania, served as a comparison community.

Observations of seat belt use were conducted both before and after the campaign in both Reading and in Bethlehem.  Twenty observation sites were used in each city. All sites were within the borders of the cities proper. Local roads and interstates were excluded from observation.  Local roads were excluded due to the low traffic volumes during night hours.  Interstates were excluded because drivers on interstates are less likely to have been exposed to the local program.  Observation sites were selected from three functional classes of roadway: principal arterial highways (5 sites), minor arterial roadways (11 sites), and urban collectors (4 sites).  Reading sites were randomly chosen from existing traffic count locations. The traffic count locations are used by PennDOT to establish vehicle miles traveled (VMT) data.  Sites in Bethlehem were randomly chosen from among all functional classes at a count station such that the number of each functional class of roadway was equal for both sites.

Day and night observations were conducted at the same observation sites.  Daytime observations occurred from 9 a.m. until 3:59 p.m.  Night observations took place from 9 p.m. until 3:59 a.m. Hours of the day (and night) did not differ by functional class.  That is, there was no systematic difference in time of observation for the three functional classes.  The same time of day for each site was used for the pre and post observations. 

The Reading program was successful despite the short duration, lack of paid media, and the fact that Pennsylvania is a secondary belt use enforcement State.  Night belt use increased by six percentage points; day belt use increased by three percentage points.  A slight (not significant) decrease in belt use was seen in Bethlehem during the daytime; there was no change in nighttime belt use.

There have been few published studies in the United States devoted to nighttime seat belt use.  All of these took place in the northeast.  In addition, with exception of the 2004 Connecticut nighttime seat belt measurement, observations were always made in lighted locations.  The present study, conducted in June 2005, took advantage of military-grade night vision technology to measure daytime versus nighttime belt use rates at the observation sites used for New Mexico’s official Federal belt use measurement.  The goal of these analyses is not to estimate seat belt use across the state of New Mexico but rather to examine the differences between daytime and nighttime seat belt use in several situations.